Content Analysis to Enhance Search

Abstract

Methods and apparatus for improving accuracy in media content searches are described. A video stream may be analyzed to identify one or more words associated with a media content item. The one or more words may be used to fulfill a search request and causing output of the media content item.

What is claimed is: 
     
         1 . A method comprising:
 adding, to a data structure for a media content item, one or more words, from a video stream, that reference the media content item; and   causing, based on a search request matching the one or more words of the data structure, output of the media content item.   
     
     
         2 . The method of  claim 1 , wherein the video stream comprises a broadcast media stream. 
     
     
         3 . The method of  claim 1 , further comprising:
 adding, to the data structure, one or more user utterances associated with the media content item.   
     
     
         4 . The method of  claim 1 , wherein the adding the one or more words further comprises:
 adding, to the data structure, a homophone of the one or more words that reference the media content item.   
     
     
         5 . The method of  claim 1 , further comprising:
 determining the one or more words that reference the media content item, based on at least one of:
 an audio signal, of the video stream, converted into a text transcript; 
 analyzing closed caption information included in the video stream; or 
 an on-screen graphic included in the video stream. 
   
     
     
         6 . The method of  claim 1 , wherein the one or more words that are associated with the media content item comprise at least one of:
 a name of the media content item;   a channel identifier associated with the media content item;   an on-demand provider associated with the media content item; or   a word or phrase included in the media content item.   
     
     
         7 . The method of  claim 1 , wherein the adding the one or more words comprises:
 determining, based on comparing an audio signal of the video stream to a database storing words or sounds commonly used by a media provider, the one or more words that reference the media content item.   
     
     
         8 . The method of  claim 1 , wherein the search request comprises a user voice input. 
     
     
         9 . The method of  claim 1 , wherein the data structure comprises:
 at least one n-gram based on at least two of the one or more words that reference the media content item.   
     
     
         10 . The method of  claim 1 , further comprising:
 detecting, based on one of an audio volume change or a scene change in the video stream, a video segment of the video stream;   wherein the one or more words that reference the media content item are from the video segment of the video stream.   
     
     
         11 . A method comprising:
 detecting, based on an audio volume change in a video stream, a video segment of the video stream;   comparing words in the video segment of the video stream to a listing of media content items;   determining, based on finding a match between at least one word from the video segment and an entry in the listing, a media content item identified by the entry in the listing; and   causing, based on a search request matching one or more words from the video segment, output of the media content item.   
     
     
         12 . The method of  claim 11 , wherein the causing output of the media content item comprises causing output of the media content item further based on one or more user utterances associated with the media content item. 
     
     
         13 . The method of  claim 11 , wherein the at least one word comprises at least one of:
 a name of the media content item;   a channel identifier associated with the media content item;   an on-demand provider associated with the media content item; or   a word or phrase included in the media content item.   
     
     
         14 . The method of  claim 11 , wherein the determining the media content item comprises:
 determining the media content item further based on a comparison of an audio signal of the video segment to a database storing words or sounds commonly used by a media provider.   
     
     
         15 . The method of  claim 11 , further comprising:
 adding, to a data structure for the media content item, the one or more words from the video segment.   
     
     
         16 . A method comprising:
 detecting, based on a scene change in a video stream, a video segment of the video stream;   comparing words in the video segment of the video stream to a listing of media content items;   determining, based on finding a match between at least one word from the video segment and an entry in the listing, a media content item identified by the entry in the listing; and   causing, based on a search request matching one or more words from the video segment, output of the media content item.   
     
     
         17 . The method of  claim 11 , wherein the causing output of the media content item comprises causing output of the media content item further based on one or more user utterances associated with the media content item. 
     
     
         18 . The method of  claim 11 , wherein the at least one word comprises at least one of:
 a name of the media content item;   a channel identifier associated with the media content item;   an on-demand provider associated with the media content item; or   a word or phrase included in the media content item.   
     
     
         19 . The method of  claim 11 , wherein the determining the media content item comprises:
 determining the media content item further based on a comparison of an audio signal of the video segment to a database storing words or sounds commonly used by a media provider.   
     
     
         20 . The method of  claim 11 , further comprising:
 adding, to a data structure for the media content item, the one or more words from the video segment.